Variable: any characteristic or phenomenon that can vary across organisms, situations, or environments.
Variables can be manipulated or measured.
Three categories of variables:
· Independent variable (usually manipulated)
· Dependent variable (usually measured)
· Subject variable (usually selected, may involve measurement)
Two types of definitions are usually found in empirical research papers. Conceptual definitions, which resemble dictionary definitions that refer to the general concepts, are often presented in the introduction. Operational definitions, which define traits in concrete, step-by-step physical terms, are usually presented in the section on method.
An operational definition
specifies the operational measures, which we use to define a concept or
variable. It must be presented in
great enough detail so another researcher could repeat it to replicate the
study.
Operational definitions should be provided for all declared variables. The following example does not describe the specific physical actions taken in the presence of the experimental participants.
“The stress-producing condition for the experimental group was a mild verbal threat given by the experimenter.”
Better: “In order to produce the stress-producing condition for the experimental group, a male experimenter who was dressed in a white doctor’s jacket seated the participants. He introduced himself as a medical doctor and stated that for the purposes of the experiment, ‘You will receive a mild electric shock while we measure your blood pressure.’”
Types of variables:
· Discrete: a variable that consists of separate, indivisible categories, so that no values can exist between two categories. E.g. type of disorder, major
· Continuous: a variable that has an infinite number of possible values, and can be represented by fractional or whole units. E.g. scores on a test, reaction time
Variables can also be categorized as
· Qualitative: a variable that varies in kind.
· Quantitative: A variable that varies in amount.
· Presence versus absence
· Amount of variable (how many levels)
· Type of variable (assign labels)
· Experimental manipulation (controlled adjustment of the IV)
Instructional manipulation
Event manipulation
· Individual difference manipulation (results in an ex post facto type of study)
Factors Affecting Our Choice of Variables to Manipulate or Observe
·
Research Tradition
Our variables may be similar to those used in previous studies
·
Theory
Theories may suggest certain variables to be included
·
Availability of new Techniques
Sometimes a new technique is developed, allowing us to manipulate a variable that we could not in the past
·
Availability of Equipment
The variables we manipulate may
be limited by the equipment available to us
The Dependent
Variable
Must
be sensitive enough to show variations in performance as a result of
variations in the IV.
Ceiling effect: Performance high at all levels of the IV
Floor effect: Performance low at all levels of the IV
Must be appropriately related to the IV (e.g. split-brain studies)
SELF-REPORT MEASURES
The most direct way to assess a construct, a hypothetical attribute or mechanism.
However, it is very easy to distort such measures.
Types of questions:
a. Open-ended: Respondents are asked to answer a question in their own words.
E.g. "What is
your opinion of your employer's affirmative action policies?"
b. Closed-ended with a limited set of response alternatives. A list of choices is given and they check the desired alternative.
E.g. "One of the
major reasons for getting married is to have children."
Agree _________ Disagree_______
c. Likert-type Scales: Respondents circle a choice, a number on a scale, or check a point on a line that best reflects their opinion.
E.g.
"The US Congress should increase spending on social programs."
· Make sure to avoid pitfalls in constructing questions, such as the following:
a. Double-barreled questions (that ask two different things). E.g. "Do you believe there are equal opportunities for minorities and women at your place of employment?
b. Loaded questions (that contain emotionally charged language). E.g. " To what extent do you think the values of the Ku Klux Klan are alive and well in the South?"
1 2 3 4 5
not uncertain very much
alive alive
Above question could be rephrased as:
"To what extent do you think support for racial segregation still exists in the South?
1 2 3 4 5
very weak uncertain very strong
support support
c.
Leading questions (that bias people to respond in a given way).
E.g. "I agree with the popular view that our current foreign policy
is flawed."
Better: "Our current foreign policy is flawed."